October 1999 - XEmacs-Beta - XEmacs Mailing List Archives

Re: 21.2.19 crash in ldap on Solaris 2.7 25 years, 1 month

Raymond Toy

>>>>> "Oscar" == Oscar Figueiredo <Oscar.Figueiredo(a)di.epfl.ch> writes: >>>>> "Raymond" == Raymond Toy <toy(a)rtp.ericsson.se> writes: Raymond> What exactly do you want me to confirm? Oscar> I just would like you to give a look at your man pages and Oscar> try to find something about ldap_set_option. You can also Oscar> have a look at the man page for ldap_open where the Oscar> preferred way of setting parameters should be indicated. I can confirm your results. ldap_set_option isn't documented, and ldap_open says to modify the fields in the LDAP structure directly. Oscar> First check the manpages. If you don't find anything about Oscar> ldap_set_option then you can try to undefine Oscar> HAVE_LDAP_SET_OPTION in the file eldap.c and recompile it. I'll try this in a few minutes. Ray

1 participants
0 comments

Re: gnus KILLS xemacs 25 years, 1 month

Jan Vroonhof

giacomo boffi <giacomo.boffi(a)polimi.it> writes: > i recompiled with --debug, similar crash, no lisp bt, this is the c bt > > (gdb) where > #0 0x402db111 in __kill () > #1 0x8078cb1 in fatal_error_signal (sig=11) at emacs.c:276 > #2 <signal handler called> > (gdb) What version of GDB do you have? There are patches on the gdb list to handle trampolines but they seem to be post-1.18. It would be nice if somebody could try them though. Jan P.S. You need to attach with gdb to XEmacs prior to the crash.

2 participants
1 comments

Re: 21.2.19 crash in ldap on Solaris 2.7 25 years, 1 month

Raymond Toy

>>>>> "Oscar" == Oscar Figueiredo <Oscar.Figueiredo(a)di.epfl.ch> writes: Oscar> The function ldap_set_option seems to be present in the Oscar> library but is not documented anywhere and the man page for Oscar> ldap_open recommends to set the options directly in the Oscar> LDAP structure (Raymond could you confirm this ?) What exactly do you want me to confirm? I assume you want me to try out Sun's ldap, but instead of using ldap_open, set the options directly, as if ldap_set_options doesn't exist? I can do this. (Got to move my UMICH libs out of the way first, though.) Ray

1 participants
0 comments

Re: (call-process ...) to return correct exit-status on NT 25 years, 1 month

Adrian Aichner

>>>>> "Andy" == Andy Piper <andy(a)xemacs.org> writes: Andy> I think this should probably go in 21.1 as well as the 21.2 Andy> series. andy Hello Andy, I agree, but was afraid to ask before I got any positive on the patch as such. Do you basically agree with what I did? I will submit formal patches for 21.2-b<of-the-day> and 21.1.7 once I get word. Regards, Adrian

2 participants
1 comments

Re: default athena3d dialogs are horrible 25 years, 1 month

Jan Vroonhof

giacomo boffi <giacomo.boffi(a)polimi.it> writes: > on the other side, it means also that up to 3n colors may be allocated > in place of n, and this is a potential problem in 8bit displays... Yes. And there is still plenty of them around. > now that configure searches Xaw3d and XEmacs is built against it, may > i say in the 90% of cases, what i'd like to see it's some lines in > "sample.Xdefaults", explaining the reasons of Xaw3d widgets' ugliness, > and how to obviate the problem OK, an entry in "sample.Xdefaults" would be fine. Jan

1 participants
0 comments

gnus KILLS xemacs 25 years, 1 month

Giacomo Boffi

(insert (emacs-version)) XEmacs 21.2 (beta19) "Shinjuku" [Lucid] (i586-pc-linux) of Mon Oct 11 1999 on boffi95.stru.polimi.it xemacs is out of cvs, monday morning CETDST (insert (gnus-version)) Gnus v5.6.45 (insert (shell-command "uname -a" t) Linux boffi95.stru.polimi.it 2.2.5-15 #1 Mon Apr 19 22:21:09 EDT 1999 i586 unknown recipe: start gnus, select any group, BOOM this is the lisp backtrace: this is the C backtrace: (gdb) where #0 0x402db111 in __kill () #1 0x8078cb1 in fatal_error_signal (sig=11) at emacs.c:276 #2 <signal handler called> (gdb) i'm going to recompile with --debug (left out to see if it matters in saving options' speed -- no, saving is slow) to see if i can get a better backtrace stay tuned

1 participants
1 comments

Re: calendar window fits no more in 80 cols 25 years, 1 month

Jan Vroonhof

giacomo boffi <giacomo.boffi(a)polimi.it> writes: > the standard (3 months) calendar window fits no more in a XEmacs frame > that's 81 cols wide (i guess it wouldnt fit in a 80 cols wide, too :-) It does for me. Are you running with Mule. If no, did you apply the fix for the string-width issue? Jan

3 participants
3 comments

Re: 21.2.19 crash in ldap on Solaris 2.7 25 years, 1 month

Oscar Figueiredo

>>>>> "Raymond" == Raymond Toy <toy(a)rtp.ericsson.se> writes: Raymond> Shoot, I spoke too soon. It worked when I first tried this, but all Raymond> subsequent trials crash when doing the actual ldap lookup. This is Raymond> using Sun's ldap library, and the C traceback is given below. Raymond> (Partial) C backtrace using Solaris 2.7's ldap, gcc 2.7.2.2: Raymond> #0 0xfeb16870 in _libc_kill () from /usr/lib/libc.so.1 Raymond> #1 0xb1ca4 in fatal_error_signal (sig=10) at emacs.c:276 Raymond> #2 <signal handler called> Raymond> #3 0xfecfc46c in ldap_set_option () from /usr/lib/libldap.so.3 Raymond> #4 0xd4464 in Fldap_open (host=10130900, plist=3766276) at eldap.c:314 I've been browsing the man pages on a Solaris 2.7 machine here at EPFL and from what I read, LDAP support in this version is somewhat experimental. The function ldap_set_option seems to be present in the library but is not documented anywhere and the man page for ldap_open recommends to set the options directly in the LDAP structure (Raymond could you confirm this ?) The function ldap_set_option is autodetected by our configure script and used if present. My feeling is that in Solaris 2.7 (at least in its current state) it shouldn't. So what do we do now ? Should I introduce that special case in the configure script ? Oscar

1 participants
0 comments

Re: default athena3d dialogs are horrible 25 years, 1 month

Jan Vroonhof

giacomo boffi <giacomo.boffi(a)polimi.it> writes: > i found it: > Emacs*dialog*beNiceToColormap: false > > could it be made the default for all Xaw3d, in the XEmacs src? If this is what I think it means then I think the answer should be "no" unless you can figure out that the visual has >8 bits depth. Jan

2 participants
1 comments

Re: hashtable-based marking and cleanups 25 years, 1 month

Robert Pluim

> Well, I get: > > EMACSBOOTSTRAPLOADPATH="/u3/exp/xemacs-20/src/../lisp/:/u3/exp/xemacs-20" EMACSBOOTSTRAPMODULEPATH="/u3/exp/xemacs-20/src/../modules/:/u3/exp/xemacs-20" ./temacs -batch -l /u3/exp/xemacs-20/src/../lisp/update-elc.el > make[1]: *** [dump-elcs] Segmentation fault (core dumped) > > This is gcc 2.95, under linux 2.2.5. I'm about to try again with > -fno-strict-aliasing (or whatever it's called) and see if it makes a > difference. Hmm, this was happenning in pdump_load, and I can't reproduce this now, so I suspect that it was trying to load the xemacs.dmp from before I applied your latest patch. (it's now failing to find ./xemacs, but I imagine that's normal ;-) temacs runs, and so far doesn't appear any slower to me. Robert -- The above are my opinions, and my opinions only.

2 participants
1 comments

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

XEmacs-Beta October 1999